Last updated:
Quick answer
Once a week, interview your bots, collect the standing instructions you gave them, and flag the ones that are slipping. Bots forget what you told them weeks ago; a weekly audit catches drift before it turns into wrong work.
Free PDF
Get this setup as a PDF
Confirm your email and we'll open a print-ready copy of this setup you can save as a PDF. You also get the free starter pack and one setup a week. Unsubscribe anytime.
Already a subscriber? Unlock it on this device
At a glance
All outcome figures are estimates from cited sources, not independent verification by us.
- What it is allowed to do
- Read-only
Only reads, researches, or summarizes. It never drafts outbound content or changes anything.
- Time saved (estimate)
- Not measured yet
- Difficulty
- Intermediate
- Tools needed
- Grok Bot
- Typical cost
- Template is free to add. Check your bot plan for usage limits.
Reviewed Oct 29, 2026
Basis (estimate): Not measured yet. We have not timed this.
What does this setup do?
Keep a fleet of bots honest about what they were told, with a weekly check that flags instructions that slipped or conflict.
Who is it for?
For anyone running more than a couple of bots who has noticed an old instruction quietly changing how a bot behaves.
How do you set it up?
Tools
- The Librarian Bot template by Brian Roemmele, added from the public directory
- A list of the bots you actually run, so the audit covers all of them
- One place to keep the audit notes, so last week can be compared with this week
Steps
- Add the Librarian Bot template and point it at the bots you run.
- Run the weekly interview: for each bot, ask what standing instructions it is following, in its own words.
- Compare the answers with the instructions you actually gave. Flag anything that slipped, changed, or conflicts with another instruction.
- Fix the flagged instructions at the source, in the bot itself, not only in the audit note.
- Keep the audit notes week to week, so a slow drift shows up as a trend instead of a surprise.
Example output
Illustrative. Your results will differ.
- Weekly list of standing instructions per bot, in the bots own words
- Flagged slips and conflicts, with both wordings shown
- No instruction changed by the audit itself
Where to paste this prompt
Grok Bot → paste the prompt into a chat, or save it as a routine or scheduled task if the job repeats
Copy-paste prompt
Adapt bracketed fields to your situation.
Interview each bot I run, one at a time, and list the standing instructions it is following, in its own words. Compare them with the instructions I gave, and flag any instruction that slipped, changed, or conflicts with another instruction. Do not change any bot instruction yourself. Report the flags and the exact wording on both sides, so I can fix them at the source.Try it on sample data first: Audit a bot without letting it fix things, a free 5-minute exercise that checks your AI's answer instantly.
Newsletter
Get the free starter pack
Free starter pack with five beginner copy-paste prompts after you confirm, plus one researched setup in your inbox each week. Unsubscribe anytime.
What usually goes wrong?
Where agents tend to stall on this job. More patterns in Where AI gets stuck.
- An audit that only checks one bot while the rest drift unnoticed; cover the whole fleet.
- A flagged instruction fixed in the note but not in the bot, so it slips again next week.
- An interview that paraphrases instead of quoting; ask for the wording as the bot holds it.
Which chat tools work with this prompt?
Paste the prompt into any of these assistants (free or paid tiers).
- Grok Bot
Common questions
- What does this setup do?
- Once a week, interview your bots, collect the standing instructions you gave them, and flag the ones that are slipping. Bots forget what you told them weeks ago; a weekly audit catches drift before it turns into wrong work.
- Who is it for?
- For anyone running more than a couple of bots who has noticed an old instruction quietly changing how a bot behaves.
- What does it cost to run?
- Template is free to add. Check your bot plan for usage limits.
- What usually goes wrong?
- An audit that only checks one bot while the rest drift unnoticed; cover the whole fleet. A flagged instruction fixed in the note but not in the bot, so it slips again next week.
- How much time might it save?
- Not measured yet. Not measured yet. We have not timed this.
What tips make it smoother?
- Run the audit on a schedule, not when something already broke.
- Fix at the source. An audit note alone does not change a bot.
- Watch for conflicts, not just drift. Two instructions can both be remembered and still disagree.
Where do the claims come from?
Summaries use public sources linked below, not independent verification. How setups are picked.
- Read Brian Roemmele, GrokBot.Rocks directory’s write-up (grokbot.rocks)
Brian Roemmele, GrokBot.Rocks directory: Directory checked 2026-10-29: updated October 28, 2026, tracking 7,539 public bot templates in 26 categories, refreshed every week. Its Librarian Bot entry by Brian Roemmele says: your bots forget what you told them weeks ago, and Librarian interviews your bots every week, finds the standing instructions you gave them, and flags the ones that are slipping.
Related setups
- AI chief of staff bot
One bot works as your front desk: it knows what each of your other bots does, passes jobs to them, and sends you a Monday status on your routines.
- How to give your AI bot a long-term memory with an Obsidian vault
Replace fuzzy in-chat “memory” with inspectable Markdown files: the bot appends a daily note and decision log to a Git-backed vault; you review in Obsidian; a weekly routine sorts an Inbox folder into PARA (Projects, Areas, Resources, Archive).